How do I install greta dependencies? | How do I know if greta has the right versions of Python dependencies installed? | Is there an alternative way to install Python dependencies? | I get the message: "Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2" | How do I make my results reproducible?

Installation | Installing TensorFlow | DiagrammeR | How greta works | Building a model | Data | explicit vs. automatic conversion | data structures | Variables and priors | variables without probability distributions | variables with probability distributions | variables with truncated distributions | Operations | Extract and replace | Functions | Likelihood | Defining the model | Plotting | Sampling | Tweaking the sampler

Rasters and vectors | Running raster and vector models on the same data | Prepare the example data | Raster approach | Vector approach | Comparing raster vs vector | Summarising connectivity metrics | Which approach should you use? | Raster approach trade-offs | Vector approach trade-offs | Converting between raster and vector | Analysis step-by-step | References

Why use targets for connectivity analysis? | A minimal _targets.R | What each section does | Running and inspecting the pipeline | Running | Inspecting results | Visualising the dependency graph | Example workflows

Quick start | Did it work? | How greta uses Python | When the default isn't right | Using greta offline | Prime the managed (uv) environment while online | Pin dependency versions | I already have Python or TensorFlow installed | I need specific dependency versions | What you can actually choose | TensorFlow 2.16 and 2.17 cannot be used | Building an unsupported installation on purpose | I want the conda workflow | How install_greta_deps() works | Advanced: how greta chooses a Python environment | Starting over | Troubleshooting

Declaring data mutable | Fitting the same model to different data | The dimensions must be the same as the original data | data_values() must be used with as_data_mutable() | Other applications | References

vis_dat | vis_miss | vis_compare | vis_expect | vis_cor | vis_value | vis_binary | vis_guess | vis_histogram() | Interactivity | Future work

vis_dat | vis_miss | vis_compare | vis_expect | vis_cor | vis_value | vis_binary | vis_guess | vis_histogram() | Interactivity | Future work

Common models | Linear regression | Multiple linear regression | Multiple Poisson regression | Multiple categorical regression | Multiple linear regression with LASSO prior | Hierarchical linear regression | Random intercept-slope model | Random intercept-slope model (with correlated effects) | Common Bayesian priors | Improper flat prior | Ridge prior | Exponential-normal prior | Horseshoe prior | Regularised horseshoe prior | Advanced Bayesian models | Hierarchical linear regression in general conditional formulation | Hierarchical linear regression in general marginal formulation | Bayesian neural network | Factor analysis | BUGS models | Air | data | greta code | BUGS/JAGS code | Stan code | Beetles | Stan models | Lightspeed | Eight schools | Ecological models | Logistic regression | Poisson regression | Logistic regression with error term | Multiple species modelling independently and concurrently | Multiple species with partial pooling of regression coefficients | Multiple species with sub-model for regression coefficients | Cormack-Jolly-Seber model

Example | Brief technical details | Bayesian interpretation of the GAM | Penalty matrices | References

Overview | The example data | The pipeline, step-by-step | Step 1: Create the barrier mask | Step 2: Remove habitat under barriers | Step 3: Buffer the habitat | Step 4: Fragment habitat along barriers | Step 5: Assign patch IDs | Step 6: Summarise patch areas | Computing connectivity metrics | Run the pipeline as a single step | Comparing multiple interpatch distances | Summary

Interpatch distance | Buffering happens on a grid | A sub-cell radius does nothing - and habitat_buffer() warns | Why it matters: the same distance, two resolutions | Choosing a resolution | What about vector (sf) data?

Accessing age and population information | Brief example of using accessor functions | An example use from the package | Using these as S3 methods in an R package | Conclusion

World data | Australian Bureau of Statistics (ABS) data | Accessing Functions | abs_age_lga() | abs_age_state() | ABS data | Education by state data for 2006 -2020 | Education by state data for 2020 | Employment by LGA for 2016 | Number of people in each household by LGA for 2016 | LGA age population for 2016 for all states and LGAs | LGA age population for 2020 for all states and LGAs | State age population for 2020 | Epidemiology / disease modelling data | Transmission probabilities from Eyre

Create a new synthetic matrix from all POLYMOD data | Generating a Next Generation Matrix | Applying Vaccination Rates | Fitting a new model with asymmetric terms

Quick example using Australian data | Quick example using world data | What next? | A More in depth example | Predicting the contact rate | Plotting | Note | Applying the model across all settings. | Fit to all settings | Predict to all settings

Introduction: What is an SIR model? | An SIR Model with homogenous mixing | Comparison to other age matrices | Calculating reproductive number - R0

Creating a next generation matrix (NGM) | Applying vaccination to an NGM

extrapolate polymod | For interest's sake: visualising the empirical contact rate data

How to run targets examples from vignettes | tar_terra_rast(): targets with terra rasters | Raster metadata | tar_terra_vect(): targets with terra vectors | tar_terra_sprc(): targets with terra raster collections | tar_terra_sds(): targets with terra raster datasets

How to run targets examples from vignettes | tar_terra_rast(): targets with terra rasters | Raster metadata | tar_terra_vect(): targets with terra vectors | tar_terra_sprc(): targets with terra raster collections | tar_terra_sds(): targets with terra raster datasets

What is an extent? | Helper functions to create multiple extents of a raster | tile_n() | tile_grid() | tile_blocksize() | How to run targets examples from vignettes | Example targets pipeline

What is an extent? | Helper functions to create multiple extents of a raster | tile_n() | tile_grid() | tile_blocksize() | How to run targets examples from vignettes | Example targets pipeline

Package Author's Notes | Indemnity Statement: | Import data from the BITRE website into R | Variables available | Crashes | Fatalities | Plot crashes by year | Plot crashes by year and state | Fatalities by year

Getting started | Exploring patterns with UpSetR | Exploring Missingness Mechanisms | geom_miss_point | General visual summaries of missing data | Missingness in variables with gg_miss_var | Missingness in cases with gg_miss_case | Missingness across factors with gg_miss_fct | Missingness along a repeating span with gg_miss_span | gg_miss_case_cumsum | gg_miss_var_cumsum | gg_miss_which

Introduction | How do we start looking at missing data? | vis_dat | vis_miss | Exploring missingness relationships | Visualising missings in variables | Replacing existing values with NA | Tidy Missing Data: The Shadow Matrix | Visualising imputed values | Numerical summaries of missing values | Using group_by with naniar | Modelling missingness | Summary | Future development | Thank you | References

Terminology | Recoding missing values

Calculating features | Creating your own Features | Accessing sets of features | Registering a feature in a package

Specify your own summaries for keys_near | Implementation of keys_near

Defining longitudinal data as a tsibble | Converting your longitudinal data to a time series | example data: wages | example: heights data | example: gapminder | example: PISA data | Conclusion

Exploring raw data | Select a sample of individuals | Filter only those with certain number of observations | Clever facets: facet_strata | Clever facets: facet_sample | Clever facets with number of observations | Exploring data using features | Plot monotonic individual series | Plot individuals with negative slope | Move along features with facet_strata | Visualise along slope

Find keys near other summaries with keys_near()

How to provide your own colour palette?

How to provide your own colour palette?

Imputing and tracking missing values | Using impute_below | Using impute_mean | Track imputed values using nabular data | Imputing values using simputation | Improving imputations | Other imputation approaches | Hmisc aregImpute

Example data | Using replace_with_na | Extending replace_with_na | Using replace_with_na_all | replace_with_na_at | replace_with_na_if | Notes on alternative ways to handle replacing with NAs

Gaussian processes in greta | Example

The data | tsibble data | Modelling | Proposed workflow

Setting up your data | Basic summaries of the data | How many observations are there? | add_n_obs() | Efficiently exploring longitudinal data | sample_n_keys() | Filtering observations | Clever facets: facet_strata | Clever facets: facet_sample | Exploratory modelling | Find keys near other summaries with keys_near | Finding features in longitudinal data | Linking individuals back to the data

Performing cross validation on max_coverage

The moviation: Why maxcovr | The problem | using max_coverage | Interpreting results | Other graphics options | Applying this to the coverage data | Future work